Fair Influence Maximization in Social Networks Based on Graph Embedding and Graph Neural Networks

As a variant of Influence Maximization (IM), Fair Influence maximization (fair IM) primarily enhances the fair propagation of information. Fair IM refers to achieving fairness in influence across the network, subject to fairness constraints, when there are initially activated nodes. It has gained increasing attention in academia. However, existing fair IM methods have difficulty ensuring fair influence on users with respect to sensitive attributes (such as race or gender), while maintaining a high distribution of influence. To address this issue, we propose a fair IM method based on graph embedding and graph neural networks, named FSGNN. In FSGNN, we design a method based on graph embedding and graph neural networks to identify potential nodes. It has a fair selection algorithm based on the proposed node selection strategy to choose seed nodes. Finally, we validate the proposed algorithm through experiments conducted on both real-world and synthetic networks. The experimental results show that the proposed method outperforms other baseline methods.